374 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



that the difference in their temperatures will become less; 

 hence a thermometer reading which exhibits great differ- 

 ence will always be earlier than one which exhibits small 

 difference. Any physical process in which a reading of this 

 kind is significant must be called an irreversible process. 

 Examples of such phenomena were listed in Chapter XIV. 1 

 In situations of this kind, at least, the directional feature 

 of time must be retained. 



Hence the state of causal laws in science seems, at present, 

 to be about as follows: Science recognizes the distinction 

 between two types of process — reversible and irreversible — 

 both of which seem to be exemplified in nature. Reversible 

 correlations are capable of adequate representation in sci- 

 ence through functional relations in which t occurs as an 

 independent variable, indifferently positive or negative. 

 Irreversible correlations are incapable of such represen- 

 tation, and require description in the general form S 2 — 

 Si > 0, where Si and S 2 represent measured values indicative, 

 respectively, of an earlier and a later state of a certain closed 

 system. If S 2 — Si = the process is reversible. If S 2 — 

 Si < the process represents an "unbecoming," hence is not 

 empirically found. Whether or not this difference in scientific 

 law represents adequately the difference in empirical proc- 

 esses, cannot perhaps, at this stage in the development of 

 science, be decided. Only two remarks may be made. In 

 the first place, there is no certainty that the fact of entropy 

 is the essence of irreversible processes; hence there is no 

 assurance that in the scientific formulation of irreversible 

 processes the important feature of causal sequences has been 

 retained. In the second place, the notion of causal efficacy 

 has been lost in the scientific formulation, unless by this 

 awareness of the immanence of the effect in the cause one 

 means simply the temporal irreversibility of the association. 

 These remarks are merely suggestive of further problems. 



(3) Scientific laws state repeated associations of events. 

 As was shown in the operational derivation, the question as 



1 Page 289. 



